Snapshot-Instantaneous ∥•∥∞ Normalization Against Heavy-Tail Noise

被引：0
|
作者：
He, Jin [1 ]
Liu, Zhong [1 ]
Wong, Kainam Thomas [2 ]
机构：
[1] Nanjing Univ Sci & Technol, Dept Elect Engn, Nanjing 210094, Jiangsu, Peoples R China
[2] Hong Kong Polytech Univ, Elect & Informat Engn Dept, Kowloon, Hong Kong, Peoples R China
来源：
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS | 2008年 / 44卷 / 03期
关键词：
D O I：
暂无
中图分类号：
V [航空、航天];
学科分类号：
08 ; 0825 ;
摘要：
A new data-dependent zero-memory preprocessing algorithm is proposed to mitigate additive heavy-tail noise and/or shot noise. Each sensor-array snapshot's spatial data vector is adaptively normalized by its infinity-norm. Zero cross-correlation is preserved between the signal-subspace and the noise-space.
引用
收藏
页码：1221 / 1227
页数：7
相关论文
共 50 条
  • [1] Robust regression with asymmetric heavy-tail noise distributions
    Takeuchi, I
    Bengio, Y
    Kanamori, T
    NEURAL COMPUTATION, 2002, 14 (10) : 2469 - 2496
  • [2] DISTRIBUTED RECURSIVE ESTIMATION UNDER HEAVY-TAIL COMMUNICATION NOISE*
    Jakovetic, Dusan
    Vukovic, Manojlo
    Bajovic, Dragana
    Sahu, Anit Kumar
    Kar, Soummya
    SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 2023, 61 (03) : 1582 - 1609
  • [3] The Heavy-Tail Phenomenon in SGD
    Gurbuzbalaban, Mert
    Simsekli, Umut
    Zhu, Lingjiong
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 139, 2021, 139
  • [4] On the foundations of multivariate heavy-tail analysis
    Resnick, S
    JOURNAL OF APPLIED PROBABILITY, 2004, 41A : 191 - 212
  • [5] Heavy-Tail Phenomenon in Decentralized SGD
    Gurbuzbalaban, Mert
    Hu, Yuanhan
    Simsekli, Umut
    Yuan, Kun
    Zhu, Lingjiong
    IISE TRANSACTIONS, 2024,
  • [6] On the Heavy-Tail Behavior of the Distributionally Robust Newsvendor
    Das, Bikramjit
    Dhara, Anulekha
    Natarajan, Karthik
    OPERATIONS RESEARCH, 2021, 69 (04) : 1077 - 1099
  • [7] Aggregation of Dependent Risks with Heavy-Tail Distributions
    Guillen, Montserrat
    Maria Sarabia, Jose
    Prieto, Faustino
    Jorda, Vanesa
    INTERNATIONAL JOURNAL OF UNCERTAINTY FUZZINESS AND KNOWLEDGE-BASED SYSTEMS, 2019, 27 (Suppl.1) : 77 - 88
  • [8] Two-Stream Multiplicative Heavy-Tail Noise Despeckling Network With Truncation Loss
    Cheng, Li
    Guo, Zhichang
    Li, Yao
    Xing, Yuming
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
  • [9] Estimating the heavy-tail index for WWW traces
    Ramirez Pacheco, J. C.
    2006 3rd International Conference on Electrical and Electronics Engineering, 2006, : 365 - 368
  • [10] When do heavy-tail distributions help?
    Hansen, Nikolaus
    Gemperle, Fabian
    Auger, Anne
    Koumoutsakos, Petros
    PARALLEL PROBLEM SOLVING FROM NATURE - PPSN IX, PROCEEDINGS, 2006, 4193 : 62 - 71
12345